[created 10/03/2012]
James Honaker
CHEM 767
December 07, 2012
Abstract
Cubanes, or cubic-shaped arrangements of carbon atoms, are a fairly recent discovery. In the five decades since their initial synthesis and discovery, the potential of highly-nitrated cubanes, such as octanitrocubane, as explosive agents has driven efforts to synthesize these compounds, and the applications outside of mere scientific curiosity for these compounds were investigated. Initial efforts were largely fruitless, though successful in the end. Properties of the nitrocubanes are largely in-line with predicted values, though the density of octanitrocubane is a notable anomaly that is the current focus of experimentation and investigation. Besides that, cubanes show promise in nanoarchitecture, as building-blocks for liquid crystals, and in pharmaceutical applications, particularly as a treatment for HIV.
Introduction and applications of cubane
Cubanes are C8 molecules arranged into a cubic formation of the carbon atoms, with one substituent off of each carbon [1]. The basic cubane molecule, C8H8 (pictured in Figure 1), was first synthesized by Philip Eaton in 1964 [2,3], the result of a synthesis process that began with 2-cyclopentenone and eventually produced the desired carbon skeleton (a process detailed in Figure 2).
Figure 1. The structure of cubane. From [1].
Figure 2. Phillip Eaton’s synthesis of cubane. From [3].
The synthesis is, as shown, a twenty-step procedure of no trivial difficulty, involving the bromination of the starting material and eventually substituting the desired building blocks. The yield of the product was low, about 30% [2,3], but it was a necessary stepping-stone towards being able to investigate properties of the system, one that initially was thought to merely be a curiosity and of little to no practical import. In stark contrast to this initial evaluation of the practicality of the synthesis and of the molecule itself, the thermodynamics, reactivities, and energetic of cubanes are of great interest, particularly the more substituted cubanes [4, 5].
On paper, cubane should not be a particularly stable molecule: the carbon-carbon bonds are at 90˚ angles to one another [1], which generates immense amounts of strain on the ring from the bonds being far from the 109.5˚ ideal usually seen in sp³ hybridized carbons [6]. If the ring ends up being broken, the reaction is strongly exothermic. However, in spite of the strain on the system, cubane is surprisingly stable, being resistant to shock and lacking routes for decomposition; the large number of bonds present, coupled with the rigid structure, means that, in spite of the strain the system experiences, it has no products it can readily break down to. It will, though, break apart energetically in the presence of oxidizers or strong bases. The molecule does melt at about 130-131˚ C [2,7], and it is soluble in a number of organic solvents (CCl4 and CHCl3 are solvents it is compatible with, but it is actually insoluble in cyclohexane). Cubane does sublimate rather readily, necessitating keeping it out of conditions where it can readily evaporate away. [7]
Before delving into one particular aspect of cubanes, their use in explosives and munitions, their potential in medical applications and in materials science should be noted. Cubanes are non-reactive in vivo, and one cubane derivative called dipivaloylcubane has shown promise as a safe treatment for HIV, acting against the virus without damaging the cells in the body. The rigidity of the cubane skeleton is also thought to be potentially assetial in the construction of nanoarchitecture and liquid crystals; polycubyl rods, which are strings of cubanes connected at diagonals, show promise in being used as a building block and as a model for certain systems, like cell bilayers [4,5].
Of particular interest is the density of cubanes. Cubanes are among the densest hydrocarbons; unsubstituted cubane has a density of 1.29 g/cm³ [8], substantially higher than the density of compounds like benzene. The density of a molecule is important to the efficiency of it as a fuel or as an explosive, as denser molecules allow for a tighter packing and a more effective deflagration system [1].
Historically, nitro groups (NO2) are conducive to effective explosives, with some other groups such as the azides (molecules with N3 groups) capable of explosive decomposition [9]; however, azide groups are extremely unstable and hyper-reactive, and so nitro groups are much more valuable in explosives. It was thus hypothesized that substituting NO2 groups onto cubane, a molecule that showed promise with its high intrinsic energy, would yield an ideal explosive. Early mathematical calculations gave a predicted density of a cubane with 8 NO2 groups as between 1.9 and 2.2 g/cm3. Using this value gives a theoretical power to this octanitrocubane that is 15-30% greater than HMX [8,10], a nitroamine explosive that is the most powerful military-grade munition at the present. It is even hypothesized to be more powerful than CL-20 (pictured in Figure 3), which is an experimental nitroamine munition currently being tested, while possessing far greater stability [10]. Cubanes are shock-resistant, even after the addition of the nitro groups, being able to sustain a hammer blow without deflagration [1]; CL-20 is unstable and susceptible to shock [11,13].
Figure 3. Structure of CL-20 (hexanitrohexaazaisowurtzitane). From [14].
In terms of actual explosive power, to use the military’s technical jargon, the relative effectiveness factor (that is, the explosive power of 1 kg of the given explosive, compared to the power of 1 kg of trinitrotoluene) of octanitrocubane is 2.38, compared to 1.80 for CL-20 and 1.70 for HMX [12]. The incentive for perfecting and refining a synthesis procedure for octanitrocubane (in particular) is pretty clear, yielding an explosive more powerful than the military’s standard and the military’s experimental compounds.
An additional benefit to cubane-based explosives is the lack of dangerous byproducts and the lack of acute toxicity on its exposure; upon decomposition, the only products should be carbon dioxide and nitrogen gas. Presently, the most common military explosive is RDX, a cyclic nitroamine which is known to cause adverse side-effects upon unwitting exposure, whether in the manufacturing process or in the field [15]. A “safer” explosive, one that can be handled without needlessly exposing troops to the health hazards of the chemicals, would almost certainly be welcome.
Synthesis of nitrocubanes
The synthesis of octanitrocubane, however, is not a straightforward task. Cubane is notably strained, and so any synthesis pathways would need to take note of its potential reactivity; adding the wrong group or adding a group at the wrong position would result in an irreversible cleaving of the cubane skeleton. The process of adding nitro groups is thus not particularly straightforward, especially if the nitro groups are adjacent to one another; attempting to substitute on groups to replace with nitro groups runs the risk of the strained carbon-carbon bond cleaving [8]. The typical synthesis pathway, then, is to start with the synthesis of the tetranitrocubane (1,3,5,7-tetranitrocubane), attaching four non-adjacent nitro groups in one convenient synthesis, and add the other nitro groups onto the nitrated skeleton in a largelt stepwise fashion [8,10].
The starting point in the synthesis of the precursor tetranitrocubane is to begin with a carboxylic acid group attached in the place of one of the hydrogens, and to convert the system to an acid chloride by the addition of COCl2; the result is an acid chloride group at the 1, 3, 5, and 7 positions. Following that, the system was treated with trimethylsilylazide (TMSN3, or (CH3)3SiN3), converting the acid chloride groups to acyl azide groups; unfortunately, azides are notoriously sensitive and are highly explosive, necessitating a good amount of caution and the use of a heavily diluted solution to lessen the risk that the primary explosive goes off and wrecks the reaction. The addition of a catalytic amine (like pyridinum hydrochloride) helped the speed of the reaction and allowed for the rearrangement of the acyl azide groups to isocyanate groups without risking the deflagration of the system. The final step was to convert the groups to nitro groups by the addition of dimethyldioxirane in wet acetone; the result was the 1,3,5,7-tetranitrocubane, a stable (shock-insensitive) compound with a density of 1.814 g/cm³ [8,16]. The hydrogens on tetranitrocubane, notably, are very acidic, especially when compared to hydrocarbons of a similar nature [8] (in general, cubane hydrogens are pretty acidic [7]). While tetranitrocubane is certainly interesting, the more-substituted nitrocubanes are more appealing, as tetranitrocubane lacks the explosive properties of the denser relatives.
From this point onwards, addition of the nitro groups has to be done through an entirely different reaction scheme. If one were to start with more carboxyl groups in the interest of spurring the additional groups that way, the only result is the cleavage of the cubane skeleton; the intermediate products would include adjacent groups where one was electron-donating and the other was electron-receiving, and that would spur the cleavage of the unstable C8 structure. The solution is a low-temperature (-78˚C) reaction, generating a monoanion with a NaN((CH3)3Si)2 salt in a 1:1 solution of tetrahydrofuran and α-methyltetrahydrofuran. Once the monoanion is formed, the system is cooled to around -125˚C, and excess N2O4 in cold isopentane is introduced, allowing the system to react and (ideally) nitrate. Following that, cold nitric acid in diethyl ether is added to quench the reaction.
This reaction does show successful nitration, and the extent of the nitration is dependent on the amount of the NaN((CH3)3Si)2 initially introduced. If there are about 4 equivalents of the salt added at the beginning, the system shows almost complete transfer to the heptanitrocubane form; the pentanitrocubane and heptanitrocubane forms dominate at lower concentrations of the salt. However, octanitrocubane is not a product from this reaction, no matter how much salt or N2O4 is added [10].
To get to octanitrocubane, the research team conducting the experiment introduced excess nitrosyl chloride to a solution of the lithium salt of the heptanitrocubane at -78˚C and ozonated it until the solution turned blue. The product from this oxidation was, indeed, the sought-after octanitrocubane, obtained at about a 50% yield [10,17,18]; its structure is given as Figure 4. The intermediate before the production of the octanitrocubane was unstable, and it was hypothesized to be nitrosoheptanitrocubane [10].
Figure 4. Structure of octanitrocubane. From [19].
Properties of nitrocubanes
While there exists very little octanitrocubane to experiment on, owing to the low yield and the recent nature of the synthesis (it was reported as successful in 2000), some basic properties have still been gathered on it and heptanitrocubane, as well as the “lesser” nitrocubanes; as mentioned before, density is of import to explosive compounds, and so densities have extensively been measured, largely through X-ray analysis [17]. As mentioned before, tetranitrocubane has a density of 1.814 g/cm³ [8]; pentanitrocubane has a density of 1.96 g/cm³, confirming a predicted trend where increased nitration results in greater density. The heptanitrocubane has an impressively high density, 2.028 g/cm³ [17]; weak hydrogen bonding is facilitated by the lone hydrogen, and intermolecular contacts between the molecules end up driving the density even higher, allowing for a more close-packed structure. Octanitrocubane, though, actually has a density that ends up being lower than predicted; theoretical and computation-assisted calculations predict a density of at least 2.123 g/cm³, but early experimental results give a density of 1.979 g/cm³ [10,17]; the same situation is present in the less-interesting hexanitrocubane, having a lower density than the now seemingly-anomalous pentanitrocubane. A chart of the densities of nitrocubanes is given in Figure 5.
Figure 5. Density of nitro-substituted cubanes. From [5].
The density of octanitrocubane is still high for a hydrocarbon, but it is lower than the density of heptanitrocubane and lower than initial hypotheses implied that it would end up [5,17]; the discrepancy is hypothesized to be partially due to the repulsions of the oxygen atoms and the inability, consequently, to achieve a packing that is as tight as heptanitrocubane gives. It is possible that the crystal form isolated is a less dense analogue of octanitrocubane (there is precedent for a heavier crystal form, as the first isolated form of CL-20 was not its densest iteration), but if there ends up being no “heavier” counterpart, it is entirely possible that the practical efficiency of octanitrocubane is less than heptanitrocubane, and that the latter ends up the more effective explosive. With that being said, octanitrocubane does have the superior explosive velocity, given as 10100 m/s. This is the highest measured explosive velocity on a non-nuclear explosive [19].
As was mentioned earlier, cubanes have a slight acidic character to them. This means that they decompose or deflagrate, sometimes violently, in the presence of bases and oxidizers; pyridine, in particular, is very effective at spurring the breakdown of cubane, and something as simple as NaF in solution can cause violent decomposition [9].
Current problems with nitrocubanes, and future avenues of research
All of the research on the nitrocubanes, though, has to date largely yielded results that have little practical application at the present time, owing to a number of complicating factors. As elucidated, the synthesis of nitrated cubanes (and, for that matter, cubane itself) is a long, painstaking process that involves multiple steps, potentially dangerous intermediates, and extremely low temperatures at specific points in the synthesis; there is no current avenue to cheaply and efficiently produce cubane or its nitrated derivatives. This leaves the costs of such materials prohibitively high, and it may be a good many years or decades before nitrocubanes see use in a practical setting, as the synthesis of a volume of nitrocubane to be used in a commercial or military setting would be a major expenditure in both time and finance.
As was shown earlier, a major impediment comes in the low yields of the products. Eaton’s initial synthesis has a 30% yield [2] that is distressingly low and a 20-step mechanism that is extremely time-consuming. There are at least two newer syntheses for cubane, though, that have emerged in the years since Eaton’s initial synthesis. One pathway begins with pentanone and reacts the system with glycol and bromine before reducing the system down and stripping away the substituent groups; this reaction has a 25% yield, though with fewer steps than the other schematic: there are only five steps in the synthesis. The other begins with 2,5-dibromo-1,4-benzoquinone and reacts with a cyclobutadiene species to produce a dicarboxylic acid, which is then reduced to the cubane (no yield was given for this synthesis by this source) [20]. The nitrocubanes have better yields for their syntheses (though the final step to obtain octanitrocubane only has about a 50% yield), but the fact remains that the reaction to even get to the starting material is incredibly inefficient. The lack of yield in the final products continues to make cubane chemistry a particularly expensive and time-consuming endeavor.
The additional target of research is the major discrepancy between the predicted density of octanitrocubane and the measured value. The question is if the crystalline form measured from the initial experiment is the only form, or if there are other crystalline structures that are denser and thus more in-line with the theoretical calculation, as is the case with CL-20 [17]. Hexanitrocubane has a similar problem, with a density that comes in below that of the 5-nitro and 7-nitro analogs (see Figure 5), though the necessity of investigating that molecule is arguably lessened in comparison to the desired heptanitrocubane and octanitrocubane. And, of course, experimental data to figure out which of the two nitrocubanes is more effective is a given. There needs to be a sufficient quantity of each compound to test, though, and that is a problem for reasons stated.
One additional compound of interest that is a homolog to cubane is the hypothetical molecule octaazacubane, an N8 molecule with similar geometry: that is, it would be a cubic formation of nitrogen atoms. This molecule has yet to be synthesized, though it is hypothesized to possess similar stability to cubane; its explosive potential, though, exceeds that of octanitrocubane. It is more dense, and it is predicted to have a higher explosive velocity [20]. Octaazacubane, though, has not been successfully synthesized, and it remains a largely hypothetical compound at this point: it would have similar properties to octanitrocubane insofar as its stability, but there has not yet been a successful synthesis of the compound to date.
Conclusion
The promise of nitrocubanes is certainly tangible at the present; they have been synthesized on at least a small scale, and they have largely been shown to have properties in line with what models and predictions have given. The next step, then, would be to test the molecules to see if they have the physical properties that prediction has suggested that they have: that is, if they possess the same explosive velocity and similar destructive potential. Considering the small amounts that have been synthesized to date thus far, it remains an ongoing process. Their medicinal properties, though only briefly touched on here, are also worth investigating: any molecule with selective activity against HIV is worth multiple glances, and it is possible that other cubane derivatives could have similar therapeutic properties.
The first point to be resolved, though, would be the discrepancies in the structure and the density of octanitrocubane, as well as if heptanitrocubane, with its greater density, is actually the desired explosive agent. Once that is concluded, the actual efficacy of the explosive can be concluded, and potential syntheses that are more efficient and expedient can commence. It could be that there is never a synthesis of octanitrocubane or heptanitrocubane that elevates the compound out of the lab and into the real world, but such efforts are certainly worth investigating. The other question would concern the synthesis of octaazacubane, the eight-nitrogen cube with even more potential power to it: could a synthesis be possible, and where would the ideal starting point of such an endeavor actually be?
And even if nitrated cubanes end up being little more than a curiosity and a fact relegated to the annals of science, cubanes still show promise in less destructive applications. Cubanes are interesting molecules, the full extent and value of which has not likely been revealed. Their uses in nanoarchitecture and pharmaceutics alone are worth pursuing research on them over. While their expense is likely to be a limiting factor for the near and foreseeable future, more efficient and effective synthetic pathways are probable, which would drive the price down to a figure more conducive to extended research.
References
[1]“Cubane”. Wikipedia. Accessed 30 November 2012. link
[2]The Cubane System. Philip E. Eaton and Thomas W. Cole J. Am. Chem. Soc.; 1964; 86(5) pp 962 - 964; DOI
[3]“Eaton Cubane Synthesis.” SynArchive. Accessed 04 December 2012. link
[4]“Cubanes.” University of Bristol. 07 December 2004. Accessed 04 December 2012. link
[5]“Cubanes”. Imperial College. Accessed 05 December 2012. link
[6] “Orbital hybridisation”. Wikipedia. Accessed 07 December 2012. link
[7] Kinetic acidity of cubane. Richard E. Dixon, Andrew Streitwieser, Philip G. Williams, and Philip E. Eaton. Journal of the American Chemical Society, 1991, 113 (1), 357-358. DOI
[8]Synthesis and Chemistry of 1,3,5,7-Tetranitrocubane Including Measurement of Its Acidity, Formation of o-Nitro Anions, and the First Preparations of Pentanitrocubane and Hexanitrocubane. Kirill A. Lukin, Jianchang Li, Philip E. Eaton, Nobuhiro Kanomata, Jürgen Hain,†, Eric Punzalan,† and, and Richard Gilardi. J. Am. Chem. Soc.; 1997; 119 (41) pp 9591-9602. DOI
[9] “Information on Azide Compounds.” Stanford Environmental Health and Safety. 08 December 2008. Accessed 07 December 2012. link
[10]Zhang, M.-X., Eaton, P. E. and Gilardi, R. (2000), Hepta- and Octanitrocubanes. Angew. Chem. Int. Ed., 39: 401–404. DOI
[11] Simpson, R. L., Urtiew, P. A., Ornellas, D. L., Moody, G. L., Scribner, K. J. and Hoffman, D. M. (1997), CL-20 performance exceeds that of HMX and its sensitivity is moderate. Propellants, Explosives, Pyrotechnics, 22: 249–255. DOI
[12] “Relative effectiveness factor”. Wikipedia. Accessed 2 December 2012. link
[13] Yirka, Bob. “University chemists devise means to stabilize explosive CL-20.” Phys.org. 09 September 2011. Accessed 05 December 2012. link
[14] “Hexanitrohexaazaisowurtzitane”. Wikipedia. Accessed 2 December 2012. link
[15] Faust, Rosemarie. “Toxicity summary for Hexacyclo-1,3,5-trinitro-1,3,5-triazine (RDX).” Oak Ridge Reservation Environmental Restoration Program. December 1994. Accessed 5 December 2012. link
[16] Agrawal, Jai, and Robert Hodgson. “Cubanes”. Organic Chemistry of Explosives. John Wiley & Sons: 2007. link to Google book
[17] Eaton, Philip E., Zhang, M.-X., Gilardi, R., Gelber, N., Iyer, S. and Surapaneni, R. (2002), Octanitrocubane: A New Nitrocarbon. Propellants, Explosives, Pyrotechnics, 27: 1–6. DOI
[18] “Octanitrocubane: Easier said than done.” The University of Chicago News Office. 20 March 2001. Accessed 7 December 2012. link
[19] “Octanitrocubane”. Wikipedia. Accessed 05 December 2012. link
[20] “Synthesis.” Imperial College. Accessed 06 December 2012. link
James Honaker
CHEM 767
December 07, 2012
Abstract
Cubanes, or cubic-shaped arrangements of carbon atoms, are a fairly recent discovery. In the five decades since their initial synthesis and discovery, the potential of highly-nitrated cubanes, such as octanitrocubane, as explosive agents has driven efforts to synthesize these compounds, and the applications outside of mere scientific curiosity for these compounds were investigated. Initial efforts were largely fruitless, though successful in the end. Properties of the nitrocubanes are largely in-line with predicted values, though the density of octanitrocubane is a notable anomaly that is the current focus of experimentation and investigation. Besides that, cubanes show promise in nanoarchitecture, as building-blocks for liquid crystals, and in pharmaceutical applications, particularly as a treatment for HIV.Introduction and applications of cubane
Cubanes are C8 molecules arranged into a cubic formation of the carbon atoms, with one substituent off of each carbon [1]. The basic cubane molecule, C8H8 (pictured in Figure 1), was first synthesized by Philip Eaton in 1964 [2,3], the result of a synthesis process that began with 2-cyclopentenone and eventually produced the desired carbon skeleton (a process detailed in Figure 2).Figure 1. The structure of cubane. From [1].
Figure 2. Phillip Eaton’s synthesis of cubane. From [3].
The synthesis is, as shown, a twenty-step procedure of no trivial difficulty, involving the bromination of the starting material and eventually substituting the desired building blocks. The yield of the product was low, about 30% [2,3], but it was a necessary stepping-stone towards being able to investigate properties of the system, one that initially was thought to merely be a curiosity and of little to no practical import. In stark contrast to this initial evaluation of the practicality of the synthesis and of the molecule itself, the thermodynamics, reactivities, and energetic of cubanes are of great interest, particularly the more substituted cubanes [4, 5].
On paper, cubane should not be a particularly stable molecule: the carbon-carbon bonds are at 90˚ angles to one another [1], which generates immense amounts of strain on the ring from the bonds being far from the 109.5˚ ideal usually seen in sp³ hybridized carbons [6]. If the ring ends up being broken, the reaction is strongly exothermic. However, in spite of the strain on the system, cubane is surprisingly stable, being resistant to shock and lacking routes for decomposition; the large number of bonds present, coupled with the rigid structure, means that, in spite of the strain the system experiences, it has no products it can readily break down to. It will, though, break apart energetically in the presence of oxidizers or strong bases. The molecule does melt at about 130-131˚ C [2,7], and it is soluble in a number of organic solvents (CCl4 and CHCl3 are solvents it is compatible with, but it is actually insoluble in cyclohexane). Cubane does sublimate rather readily, necessitating keeping it out of conditions where it can readily evaporate away. [7]
Before delving into one particular aspect of cubanes, their use in explosives and munitions, their potential in medical applications and in materials science should be noted. Cubanes are non-reactive in vivo, and one cubane derivative called dipivaloylcubane has shown promise as a safe treatment for HIV, acting against the virus without damaging the cells in the body. The rigidity of the cubane skeleton is also thought to be potentially assetial in the construction of nanoarchitecture and liquid crystals; polycubyl rods, which are strings of cubanes connected at diagonals, show promise in being used as a building block and as a model for certain systems, like cell bilayers [4,5].
Of particular interest is the density of cubanes. Cubanes are among the densest hydrocarbons; unsubstituted cubane has a density of 1.29 g/cm³ [8], substantially higher than the density of compounds like benzene. The density of a molecule is important to the efficiency of it as a fuel or as an explosive, as denser molecules allow for a tighter packing and a more effective deflagration system [1].
Historically, nitro groups (NO2) are conducive to effective explosives, with some other groups such as the azides (molecules with N3 groups) capable of explosive decomposition [9]; however, azide groups are extremely unstable and hyper-reactive, and so nitro groups are much more valuable in explosives. It was thus hypothesized that substituting NO2 groups onto cubane, a molecule that showed promise with its high intrinsic energy, would yield an ideal explosive. Early mathematical calculations gave a predicted density of a cubane with 8 NO2 groups as between 1.9 and 2.2 g/cm3. Using this value gives a theoretical power to this octanitrocubane that is 15-30% greater than HMX [8,10], a nitroamine explosive that is the most powerful military-grade munition at the present. It is even hypothesized to be more powerful than CL-20 (pictured in Figure 3), which is an experimental nitroamine munition currently being tested, while possessing far greater stability [10]. Cubanes are shock-resistant, even after the addition of the nitro groups, being able to sustain a hammer blow without deflagration [1]; CL-20 is unstable and susceptible to shock [11,13].
Figure 3. Structure of CL-20 (hexanitrohexaazaisowurtzitane). From [14].
In terms of actual explosive power, to use the military’s technical jargon, the relative effectiveness factor (that is, the explosive power of 1 kg of the given explosive, compared to the power of 1 kg of trinitrotoluene) of octanitrocubane is 2.38, compared to 1.80 for CL-20 and 1.70 for HMX [12]. The incentive for perfecting and refining a synthesis procedure for octanitrocubane (in particular) is pretty clear, yielding an explosive more powerful than the military’s standard and the military’s experimental compounds.
An additional benefit to cubane-based explosives is the lack of dangerous byproducts and the lack of acute toxicity on its exposure; upon decomposition, the only products should be carbon dioxide and nitrogen gas. Presently, the most common military explosive is RDX, a cyclic nitroamine which is known to cause adverse side-effects upon unwitting exposure, whether in the manufacturing process or in the field [15]. A “safer” explosive, one that can be handled without needlessly exposing troops to the health hazards of the chemicals, would almost certainly be welcome.
Synthesis of nitrocubanes
The synthesis of octanitrocubane, however, is not a straightforward task. Cubane is notably strained, and so any synthesis pathways would need to take note of its potential reactivity; adding the wrong group or adding a group at the wrong position would result in an irreversible cleaving of the cubane skeleton. The process of adding nitro groups is thus not particularly straightforward, especially if the nitro groups are adjacent to one another; attempting to substitute on groups to replace with nitro groups runs the risk of the strained carbon-carbon bond cleaving [8]. The typical synthesis pathway, then, is to start with the synthesis of the tetranitrocubane (1,3,5,7-tetranitrocubane), attaching four non-adjacent nitro groups in one convenient synthesis, and add the other nitro groups onto the nitrated skeleton in a largelt stepwise fashion [8,10].The starting point in the synthesis of the precursor tetranitrocubane is to begin with a carboxylic acid group attached in the place of one of the hydrogens, and to convert the system to an acid chloride by the addition of COCl2; the result is an acid chloride group at the 1, 3, 5, and 7 positions. Following that, the system was treated with trimethylsilylazide (TMSN3, or (CH3)3SiN3), converting the acid chloride groups to acyl azide groups; unfortunately, azides are notoriously sensitive and are highly explosive, necessitating a good amount of caution and the use of a heavily diluted solution to lessen the risk that the primary explosive goes off and wrecks the reaction. The addition of a catalytic amine (like pyridinum hydrochloride) helped the speed of the reaction and allowed for the rearrangement of the acyl azide groups to isocyanate groups without risking the deflagration of the system. The final step was to convert the groups to nitro groups by the addition of dimethyldioxirane in wet acetone; the result was the 1,3,5,7-tetranitrocubane, a stable (shock-insensitive) compound with a density of 1.814 g/cm³ [8,16]. The hydrogens on tetranitrocubane, notably, are very acidic, especially when compared to hydrocarbons of a similar nature [8] (in general, cubane hydrogens are pretty acidic [7]). While tetranitrocubane is certainly interesting, the more-substituted nitrocubanes are more appealing, as tetranitrocubane lacks the explosive properties of the denser relatives.
From this point onwards, addition of the nitro groups has to be done through an entirely different reaction scheme. If one were to start with more carboxyl groups in the interest of spurring the additional groups that way, the only result is the cleavage of the cubane skeleton; the intermediate products would include adjacent groups where one was electron-donating and the other was electron-receiving, and that would spur the cleavage of the unstable C8 structure. The solution is a low-temperature (-78˚C) reaction, generating a monoanion with a NaN((CH3)3Si)2 salt in a 1:1 solution of tetrahydrofuran and α-methyltetrahydrofuran. Once the monoanion is formed, the system is cooled to around -125˚C, and excess N2O4 in cold isopentane is introduced, allowing the system to react and (ideally) nitrate. Following that, cold nitric acid in diethyl ether is added to quench the reaction.
This reaction does show successful nitration, and the extent of the nitration is dependent on the amount of the NaN((CH3)3Si)2 initially introduced. If there are about 4 equivalents of the salt added at the beginning, the system shows almost complete transfer to the heptanitrocubane form; the pentanitrocubane and heptanitrocubane forms dominate at lower concentrations of the salt. However, octanitrocubane is not a product from this reaction, no matter how much salt or N2O4 is added [10].
To get to octanitrocubane, the research team conducting the experiment introduced excess nitrosyl chloride to a solution of the lithium salt of the heptanitrocubane at -78˚C and ozonated it until the solution turned blue. The product from this oxidation was, indeed, the sought-after octanitrocubane, obtained at about a 50% yield [10,17,18]; its structure is given as Figure 4. The intermediate before the production of the octanitrocubane was unstable, and it was hypothesized to be nitrosoheptanitrocubane [10].
Figure 4. Structure of octanitrocubane. From [19].
Properties of nitrocubanes
While there exists very little octanitrocubane to experiment on, owing to the low yield and the recent nature of the synthesis (it was reported as successful in 2000), some basic properties have still been gathered on it and heptanitrocubane, as well as the “lesser” nitrocubanes; as mentioned before, density is of import to explosive compounds, and so densities have extensively been measured, largely through X-ray analysis [17]. As mentioned before, tetranitrocubane has a density of 1.814 g/cm³ [8]; pentanitrocubane has a density of 1.96 g/cm³, confirming a predicted trend where increased nitration results in greater density. The heptanitrocubane has an impressively high density, 2.028 g/cm³ [17]; weak hydrogen bonding is facilitated by the lone hydrogen, and intermolecular contacts between the molecules end up driving the density even higher, allowing for a more close-packed structure. Octanitrocubane, though, actually has a density that ends up being lower than predicted; theoretical and computation-assisted calculations predict a density of at least 2.123 g/cm³, but early experimental results give a density of 1.979 g/cm³ [10,17]; the same situation is present in the less-interesting hexanitrocubane, having a lower density than the now seemingly-anomalous pentanitrocubane. A chart of the densities of nitrocubanes is given in Figure 5.Figure 5. Density of nitro-substituted cubanes. From [5].
The density of octanitrocubane is still high for a hydrocarbon, but it is lower than the density of heptanitrocubane and lower than initial hypotheses implied that it would end up [5,17]; the discrepancy is hypothesized to be partially due to the repulsions of the oxygen atoms and the inability, consequently, to achieve a packing that is as tight as heptanitrocubane gives. It is possible that the crystal form isolated is a less dense analogue of octanitrocubane (there is precedent for a heavier crystal form, as the first isolated form of CL-20 was not its densest iteration), but if there ends up being no “heavier” counterpart, it is entirely possible that the practical efficiency of octanitrocubane is less than heptanitrocubane, and that the latter ends up the more effective explosive. With that being said, octanitrocubane does have the superior explosive velocity, given as 10100 m/s. This is the highest measured explosive velocity on a non-nuclear explosive [19].
As was mentioned earlier, cubanes have a slight acidic character to them. This means that they decompose or deflagrate, sometimes violently, in the presence of bases and oxidizers; pyridine, in particular, is very effective at spurring the breakdown of cubane, and something as simple as NaF in solution can cause violent decomposition [9].
Current problems with nitrocubanes, and future avenues of research
All of the research on the nitrocubanes, though, has to date largely yielded results that have little practical application at the present time, owing to a number of complicating factors. As elucidated, the synthesis of nitrated cubanes (and, for that matter, cubane itself) is a long, painstaking process that involves multiple steps, potentially dangerous intermediates, and extremely low temperatures at specific points in the synthesis; there is no current avenue to cheaply and efficiently produce cubane or its nitrated derivatives. This leaves the costs of such materials prohibitively high, and it may be a good many years or decades before nitrocubanes see use in a practical setting, as the synthesis of a volume of nitrocubane to be used in a commercial or military setting would be a major expenditure in both time and finance.
As was shown earlier, a major impediment comes in the low yields of the products. Eaton’s initial synthesis has a 30% yield [2] that is distressingly low and a 20-step mechanism that is extremely time-consuming. There are at least two newer syntheses for cubane, though, that have emerged in the years since Eaton’s initial synthesis. One pathway begins with pentanone and reacts the system with glycol and bromine before reducing the system down and stripping away the substituent groups; this reaction has a 25% yield, though with fewer steps than the other schematic: there are only five steps in the synthesis. The other begins with 2,5-dibromo-1,4-benzoquinone and reacts with a cyclobutadiene species to produce a dicarboxylic acid, which is then reduced to the cubane (no yield was given for this synthesis by this source) [20]. The nitrocubanes have better yields for their syntheses (though the final step to obtain octanitrocubane only has about a 50% yield), but the fact remains that the reaction to even get to the starting material is incredibly inefficient. The lack of yield in the final products continues to make cubane chemistry a particularly expensive and time-consuming endeavor.
The additional target of research is the major discrepancy between the predicted density of octanitrocubane and the measured value. The question is if the crystalline form measured from the initial experiment is the only form, or if there are other crystalline structures that are denser and thus more in-line with the theoretical calculation, as is the case with CL-20 [17]. Hexanitrocubane has a similar problem, with a density that comes in below that of the 5-nitro and 7-nitro analogs (see Figure 5), though the necessity of investigating that molecule is arguably lessened in comparison to the desired heptanitrocubane and octanitrocubane. And, of course, experimental data to figure out which of the two nitrocubanes is more effective is a given. There needs to be a sufficient quantity of each compound to test, though, and that is a problem for reasons stated.
One additional compound of interest that is a homolog to cubane is the hypothetical molecule octaazacubane, an N8 molecule with similar geometry: that is, it would be a cubic formation of nitrogen atoms. This molecule has yet to be synthesized, though it is hypothesized to possess similar stability to cubane; its explosive potential, though, exceeds that of octanitrocubane. It is more dense, and it is predicted to have a higher explosive velocity [20]. Octaazacubane, though, has not been successfully synthesized, and it remains a largely hypothetical compound at this point: it would have similar properties to octanitrocubane insofar as its stability, but there has not yet been a successful synthesis of the compound to date.
Conclusion
The promise of nitrocubanes is certainly tangible at the present; they have been synthesized on at least a small scale, and they have largely been shown to have properties in line with what models and predictions have given. The next step, then, would be to test the molecules to see if they have the physical properties that prediction has suggested that they have: that is, if they possess the same explosive velocity and similar destructive potential. Considering the small amounts that have been synthesized to date thus far, it remains an ongoing process. Their medicinal properties, though only briefly touched on here, are also worth investigating: any molecule with selective activity against HIV is worth multiple glances, and it is possible that other cubane derivatives could have similar therapeutic properties.The first point to be resolved, though, would be the discrepancies in the structure and the density of octanitrocubane, as well as if heptanitrocubane, with its greater density, is actually the desired explosive agent. Once that is concluded, the actual efficacy of the explosive can be concluded, and potential syntheses that are more efficient and expedient can commence. It could be that there is never a synthesis of octanitrocubane or heptanitrocubane that elevates the compound out of the lab and into the real world, but such efforts are certainly worth investigating. The other question would concern the synthesis of octaazacubane, the eight-nitrogen cube with even more potential power to it: could a synthesis be possible, and where would the ideal starting point of such an endeavor actually be?
And even if nitrated cubanes end up being little more than a curiosity and a fact relegated to the annals of science, cubanes still show promise in less destructive applications. Cubanes are interesting molecules, the full extent and value of which has not likely been revealed. Their uses in nanoarchitecture and pharmaceutics alone are worth pursuing research on them over. While their expense is likely to be a limiting factor for the near and foreseeable future, more efficient and effective synthetic pathways are probable, which would drive the price down to a figure more conducive to extended research.
References
[1]“Cubane”. Wikipedia. Accessed 30 November 2012. link[2]The Cubane System. Philip E. Eaton and Thomas W. Cole J. Am. Chem. Soc.; 1964; 86(5) pp 962 - 964; DOI
[3]“Eaton Cubane Synthesis.” SynArchive. Accessed 04 December 2012. link
[4]“Cubanes.” University of Bristol. 07 December 2004. Accessed 04 December 2012. link
[5]“Cubanes”. Imperial College. Accessed 05 December 2012. link
[6] “Orbital hybridisation”. Wikipedia. Accessed 07 December 2012. link
[7] Kinetic acidity of cubane. Richard E. Dixon, Andrew Streitwieser, Philip G. Williams, and Philip E. Eaton. Journal of the American Chemical Society, 1991, 113 (1), 357-358. DOI
[8]Synthesis and Chemistry of 1,3,5,7-Tetranitrocubane Including Measurement of Its Acidity, Formation of o-Nitro Anions, and the First Preparations of Pentanitrocubane and Hexanitrocubane. Kirill A. Lukin, Jianchang Li, Philip E. Eaton, Nobuhiro Kanomata, Jürgen Hain,†, Eric Punzalan,† and, and Richard Gilardi. J. Am. Chem. Soc.; 1997; 119 (41) pp 9591-9602. DOI
[9] “Information on Azide Compounds.” Stanford Environmental Health and Safety. 08 December 2008. Accessed 07 December 2012. link
[10]Zhang, M.-X., Eaton, P. E. and Gilardi, R. (2000), Hepta- and Octanitrocubanes. Angew. Chem. Int. Ed., 39: 401–404. DOI
[11] Simpson, R. L., Urtiew, P. A., Ornellas, D. L., Moody, G. L., Scribner, K. J. and Hoffman, D. M. (1997), CL-20 performance exceeds that of HMX and its sensitivity is moderate. Propellants, Explosives, Pyrotechnics, 22: 249–255. DOI
[12] “Relative effectiveness factor”. Wikipedia. Accessed 2 December 2012. link
[13] Yirka, Bob. “University chemists devise means to stabilize explosive CL-20.” Phys.org. 09 September 2011. Accessed 05 December 2012. link
[14] “Hexanitrohexaazaisowurtzitane”. Wikipedia. Accessed 2 December 2012. link
[15] Faust, Rosemarie. “Toxicity summary for Hexacyclo-1,3,5-trinitro-1,3,5-triazine (RDX).” Oak Ridge Reservation Environmental Restoration Program. December 1994. Accessed 5 December 2012. link
[16] Agrawal, Jai, and Robert Hodgson. “Cubanes”. Organic Chemistry of Explosives. John Wiley & Sons: 2007. link to Google book
[17] Eaton, Philip E., Zhang, M.-X., Gilardi, R., Gelber, N., Iyer, S. and Surapaneni, R. (2002), Octanitrocubane: A New Nitrocarbon. Propellants, Explosives, Pyrotechnics, 27: 1–6. DOI
[18] “Octanitrocubane: Easier said than done.” The University of Chicago News Office. 20 March 2001. Accessed 7 December 2012. link
[19] “Octanitrocubane”. Wikipedia. Accessed 05 December 2012. link
[20] “Synthesis.” Imperial College. Accessed 06 December 2012. link